EP2088722A1 - Method for transferring data and network for executing same - Google Patents

Abstract

The method involves transmitting user data representing a file or data flow in form of data packets among computer workstations (4.1-4.3), a conning tower (4.4), sensors (4.5), a camera monitoring system (4.6), and a radar equipment (4.7) over a core network (2). The user data is guided over core switching processors e.g. core-routers (3.1-3.3). The workstations, tower, the sensors, the monitoring system and the radar equipment are redundantly connected at the processors. Operating data with actual dynamic routing tables are regularly exchanged by the processors below each other. An independent claim is also included for a network for packet-switching transmission of data over an end-to-end connection.

Description

The invention relates to a method for transmitting data in a network via packet-switched end-to-end connections. It also relates to a network for carrying out the method, wherein the corresponding network according to an embodiment of the invention is implemented on board a ship.

The transmission or the acquisition of information in the form of data often takes place at the present time via networks, via which the data interchanging devices and systems are interconnected. The probably most complex network is the Internet, which is not a rigid network with fixed characteristics and spatial extent with regard to the included components, but rather a loose computer network, which however in the form of so-called backbones at least core areas with a relatively fixed structure having.

In addition to Wide Area Networks (WAN), such as the Internet, smaller and locally formed and arranged networks, that is, Local Area Networks (LAN) with a fundamentally comparable structure are known. Even with local networks, core areas of the network and access areas can be different from each other, especially if they are of a size. Thus, for example, in a LAN implemented in a multi-storey building in connection with the units interconnecting the individual floors and their connection to each other, this is also referred to as a backbone. Also, different units of technical systems, such as aircraft, motor vehicles or ships, are nowadays often networked together, that is part of more or less complex networks.

The networks used today are mostly IP-based networks. A typical IP network consists of the network end devices communicating with each other via the network and a plurality of switches and / or routers. The data transmission takes place in IP networks, unlike in the fixed telephone network, packet-switched and thus not circuit-switched. In this case, files or data streams to be transmitted are subdivided into data blocks with a format determined by the respective transmission protocol used and transmitted via the switches or routers from a network terminal device which transmits the data to a network terminal device receiving it. On the basis of the respective receiving network terminal designating destination address, the data packets through the switches and

Router the forwarding device provided for the reception.

In order to ensure a high availability of the networks, it is known to provide a redundant number of switches or routers in a core area of a network. The switches or routers determine the path of the data packets through the network on the basis of a routing table in which they map the respectively surrounding network structure with neighboring routers or switches. If a fault occurs due to the failure of one or more components or due to a temporary overload in the path provided by the network for a data packet, the determination takes place after a predetermined period of time (timeout) within which a data packet can not be forwarded a new possible way using the redundant switches or routers. The use of an alternative, implemented by redundant units way, however, takes place according to common network concepts only in the case of failure of the originally intended for data transmission path through the network. Thus, although the availability of the network can be increased by the arrangement of redundant units, but the individual connection path through the network despite using high-frequency transmission paths, such as those provided by the fiber optic technology and high-performance core switching computers (switches or routers), at a high data volumes prove to be a bottleneck. In addition, known network concepts have the disadvantage that, in the event of disturbances occurring, the time required for switching the signal path using the redundant components is comparatively high. For example, the switching time on the Internet can be up to to 60 seconds. Although this is unproblematic in many cases, it is unacceptable for time-critical applications. For example, for networks over which technical components of a motor vehicle are networked with each other, significantly shorter switching times are needed and sought in case of disturbances occurring. The same applies a fortiori to even more complex systems, such as ocean-going vessels or naval vessels, in particular if these are possibly components of military security systems. For example, for floating combat units, a switching time of less than 2 seconds is required for switching from a disturbed transmission path to a redundant transmission path relative to a Layer 4 end-to-end connection. On the other hand, according to the prior art, the situation can be observed, especially on large ships often, isolated solutions of several independent IT systems exist. However, solutions would also be desirable here in which the IT systems work in a comprehensive network with common but sometimes also different data.

The object of the invention is to provide a solution by which a high availability and performance is ensured in the data transmission. For this purpose, a procedure and the structure of a corresponding network must be specified.

The object is achieved by a method having the features of the main claim. A task-solving suitable for carrying out the process network is characterized by the first claim. Advantageous Ausbeziehungsweise developments of the invention are achieved by the dependent claims.

The method of transferring data in a network proposed for solving the problem relates to a method in which the data is transmitted via packet-switched end-to-end connections. After that, payload data representing a file or a data stream is transmitted over the network in the form of data packets. They are doing this by a transmitting the user data terminal of the network comprising a plurality of terminals via a core network with at least two trained as high performance units core switching computers to another, the user data receiving terminal transmitted. In the context of the presentation of the invention, the abovementioned terminal devices of the network are either either a network-capable terminal or a further network which, with regard to the network considered with the core network as part of the representation of the invention, is to be regarded as a subnet , The data packets of a file or of a data stream are routed via the core switching computers and a route determined by them through the network of the respective terminals provided for receiving the user data. In this case, the path through the network is determined by the central switching computers in a manner known in principle on the basis of a destination address which identifies the receiving terminal of the relevant end-to-end connection contained in a respective data packet.

According to the invention, the user data of a file or of a data stream is routed via one or more of the at least two core switching computers of the core network, but each data packet of this data only passes through one of the core switching computers. For this purpose, according to the invention, all terminal devices belonging to the network are connected redundantly to all core switching computers. Furthermore, all core exchange computers constantly exchange operating data with each other, which comprise at least their respective current dynamic routing tables.

While according to the prior art in networks of known architecture, the user data from acting as a source terminal, optionally with the interposition of a source access switch, first via a source-core router, optionally further core router and finally a target core -Router, in turn, in turn, with the interposition of a destination access switch, that is therefore supplied in each case via a plurality of core router of the intended destination terminal, the data packets are in the process according to the invention in each case only one Kernvermittlungsrechner (core router or Layer 3 switch) out. The prerequisite for this is the direct or indirect connection of all terminal devices of the network to all central switching computers. Advantageously, this results in shorter transmission paths, so that the data transmission rate increases. By constantly taking place between the core switching facilities exchange of operating data, in particular the routing tables, a very fast switching to another of the redundant data paths is also possible in case of failure of a selected transmission path.

The method according to the invention is advantageously designed such that the entire data traffic resulting from the transmission of user data in the network is dynamically distributed to the core switching computers. The distribution of the data traffic is controlled by units on or in the respective user data emitting terminal equipment. Preferably, this is done according to the principle of load balancing by means of a software routine processed by the respective units.

The method can also be further developed in that the data, depending on their nature, are forwarded by the core switching computers with different priorities. In this case, the prioritization preferably takes place on the basis of known quality-of-service mechanisms, so that the different prioritization of data, if appropriate, also comes into effect only if the quality of service is inadequate.

A further possible embodiment of the method according to the invention, in which subnetworks with components of different system areas are connected to the central switching computers of the network within a technical system, is given by the fact that the data in the network are transmitted via identical physical paths but logically separated from one another , For this purpose, virtual networks are defined in the network, which are assigned to the different system areas. The method according to the invention can also advantageously be further developed by virtue of the data being distributed throughout the network or between selected ones virtual networks, according to the previously explained embodiment, are transmitted encrypted.

The network for packet-switched transmission of data via end-to-end connections designed to solve the problem and to carry out the method according to the invention consists of a core network with at least two core switching computers designed as high-performance units, an access area surrounding the core network and a plurality of access areas the core switching computers of the core network connected terminal equipment. In this context, the characterization of the central office computers as high performance units refers to that the data transmission rate in the core network and thus also in the core switching computers is higher (preferably by a power of ten or more) than in the surrounding network areas or in the access areas. This is also the case with networks of the prior art, whereby here, at present, transmission rates of approximately 10 Gbit / s are usually to be found in the core network.

As already described in connection with the method, the terminal devices are in each case a network-capable terminal or a subnet. According to the invention, the network is designed such that each terminal device, at least during the existence of an end-to-end connection involving it, is connected to all core switching computers of the core network. Furthermore, all the central switching computers are meshed with one another, wherein between them one or more, exclusively the exchange of operating data, in particular the current dynamic routing tables, serving high-performance connections exist.

According to a possible embodiment of the network according to the invention, the core switching computers already mentioned repeatedly are layer-3 switches. In addition, it is also possible to realize the core switching computers through core routers. Of course, constellations are also conceivable in which both layer-3 switches and core computers are used. Regardless of their nature, the core switching computers are preferably equipped with a firewall. The Meshing of the central office computers is advantageously realized in that all central office computers are interconnected to exchange the operating data by one or more fiber optic cables.

As already stated earlier, the terminal equipment can be connected directly or indirectly to the central office computers. According to a possible embodiment, one or more terminals are indirectly connected to the central office computers via switches arranged in the access area of the network (usually layer 2 switches). In any case, however, it is also ensured here in accordance with the basic principle of the invention that all terminal devices are connected via the switches of the access area to all core switching computers of the core network. The connection of the terminal equipment to the central office computers can also be realized with the interposition of means for data encryption and decryption. Depending on the requirements - more detailed explanations are given in the course of the presentation of an embodiment - may be formed in the network subnets with at least two terminals and at least two core exchange computers, which are connected to the other network parts each via a security gateway.

According to an envisaged realization, the network according to the invention is implemented on board a ship. The terminal devices in this case are individual devices or subnetwork devices of navigation, ship automation and communication, sensors and effectors or actuators and devices for monitoring and controlling the aforementioned components. The core switching computers (layer 3 switches or core routers) are arranged in this embodiment of the network according to the invention with respect to the ship's longitudinal extension in different sections of the ship. They are connected to each other via at least two on each side of the ship along guided connection lines. Preferably, at least one core exchange computer in the bow area and another core exchange computer in the rear area of the ship are arranged. According to a further development, the implementation on board a ship may concern Embodiment of the network according to the invention further comprise components of the onboard air conditioning terminal equipment. If the ship is a floating combat unit or a ship of the navy, the terminal equipment included in the network may also include components of a board armament.

Fig. 1:

ein grundsätzliches Schema eines erfindungsgemäßen Netzwerks,

Fig. 2:

eine mögliche Ausbildungsform des erfindungsgemäßen Netzwerks in Form einer Implementierung an Bord eines Schiffes,

Fig. 3:

ein Netzwerk nach dem Stand der Technik.

The invention will be explained in more detail below with reference to drawings with an embodiment and compared to the prior art. In the accompanying drawings show:

Fig. 1:

a basic diagram of a network according to the invention,

Fig. 2:

a possible embodiment of the network according to the invention in the form of an implementation on board a ship,

3:

a network according to the prior art.

The Fig. 1 shows the network 1 according to the invention in the form of a basic scheme. The network 1 consists of the core network 2 with the core switching computers 3.1 - 3.3 arranged therein, the access area 5 surrounding the core network 2 and the terminals 4.1 - 4.7, which according to the exemplary representation via access switches (6.1 - 6.6) of the access area 5 to the Kernvermittlungsrechner 3.1 - 3.3, here are assumed core router, are connected. As can be seen, the terminals 4.1 - 4.7 are network-capable terminals, for example computers with a network card, and a subnet 7, for example a LAN. The connection of the terminals 4.1 - 4.7 to the core network 2 is, in accordance with the basic idea of the invention, such that each terminal 4.1 - 4.7 is connected via the respective access switch 6.1 - 6.6 to each of the core routers 3.1 - 3.3 in the core network 2 , Within the core network 2, each core router 3.1 - 3.3 is connected to each other core router 3.1 - 3.3 of the core network 2 by at least one high-performance fiber optic cable 8. The fiber optic cables 8 form a backbone, which is not included in a respective end-to-end connection with respect to the transmission of the user data, that is, over which no data packets are carried with user data, but which merely serves for the exchange of operating data between the core routers 3.1 - 3.3. In addition to possibly data on network utilization and quality, the core routers 3.1 - 3.3 constantly exchange their current routing tables via the backbone. As a result of the illustrated structure, short paths through the core network 2 are realized for end-to-end connections, so that high data transmission rates are achieved, whereby the redundant connection of the terminals 4.1 - 4.7 to the core network 2, respectively the core routers 3.1 - 3.3, a high availability of the network 1 is guaranteed. Due to the constant synchronization of the core routers 3.1-3.3 or their routing tables, in the event of a failure of an element or a device of the currently used transmission path, very rapid switching to an alternative transmission path is also made possible.

The inventive concept described above is confronted with networks 1 of the prior art having a structure as shown schematically in FIG Fig. 3 is shown. The Fig. 3 shows the structure of a classic IP network, as it is currently found. Also networks 1 of the known type exist, as shown in Fig. 3 can be seen, from a core network 2, an access area 5 and via the access area 5 to the core network 2 connected terminals 4.1 - 4.7. In contrast to the invention, in this network structure, the terminals 4.1 - 4.7 are each connected only to some of the core routers 3.1 - 3.3 in the core area or core network 2. Although the core routers 3.1 - 3.3 are also interconnected here, user data is necessarily transmitted via these connections 8. For example, the establishment of an end-to-end connection and the transmission of user data between the terminal 4.1 and the terminal 4.7 via the access switch 6.2, the core router 3.1, the core router 3.3 and the access switch 6.5 done. In the case of an interruption of the direct connection between the core routers 3.1 and 3.3, the user data in the core network 2 can alternatively be routed via the core router 3.1, the core router 3.2 and then via the core router 3.3. Although this ensures a high availability of the network 1, but takes place the entire network traffic with respect to the user data in each case involving multiple core routers 3.1, 3.2, 3.3, so that the performance compared to the network structure according to the invention is lower. In addition, since in networks of the illustrated known structure between the core routers 3.1 - 3.3 is no constant exchange or no permanent synchronization of the routing tables, also in case of the need to switch the data path, a larger switching delay is recorded.

In the Fig. 2 is again a concrete embodiment with an implementation of the network 1 according to the invention on board a ship shown in a schematic representation. At, in the example shown, two core routers 3.1 - 3.2 or Layer 3 switches of the core network 2, several terminal devices of the ship are redundantly connected, the illustration shows only some of these terminal devices by way of example. The core routers 3.1 - 3.2 are redundantly interconnected by two or more fiber optic cables 8. About this backbone 8 takes place, as already emphasized several times, only an exchange of operating data between the core routers 3.1 - 3.2. In contrast, all data packets with user data run in each case from one terminal device 4.1 - 4.7 via only one core router 3.1 - 3.2 to the respective receiving network device 4.1 - 4.7, which at most via an access switch 6.1 - 6.4 and not directly to the relevant core device. Router 3.1 - 3.2 is connected. In the example shown, a radar device 4.7, a camera monitoring system 4.6, additional sensors 4.5 and a command post 4.4 are connected directly to the core switches 3.1 - 3.2 redundantly. Computer workstations 4.1 - 4.3 for receiving and evaluating the data transmitted by the aforementioned terminal devices 4.4 - 4.7 are indirectly connected to the core routers 3.1 - 3.2 via access switches 6.1 - 6.4 of the access area. Each of the terminal devices 4.1 - 4.7 shown schematically may be a single device or a more complex system, or a subnet 7 with a plurality of identical or interacting devices. In addition, individual terminal devices 4.1 - 4.7 can be combined into virtual networks. However, this happens purely logical and not on physical level, so that ultimately all data packets with user data each run on one of the two core routers 3.1 - 3.2 of the core network. However, the logical separation makes it possible to prioritize data or to completely or partially shield data exchanged between certain virtual networks by encryption and / or inclusion of security gateways against access by other terminals 4.1 - 4.7. Thus, for example, data from radar devices 4.7 or ship navigation with a higher priority are transmitted on board a ship than, for example, data for in-flight entertainment. Corresponding security structures can be designed in such a way that data from other areas can indeed be received by devices in virtual networks of higher security levels, but that data from the networks with a high security level does not go out into the other areas of the overall network. In the example shown, the system time of the terminals 4.1 - 4.7 can be synchronized via the network 1 with a GPS time standard.

Used reference signs

1

network

2

core network

3.1 - 3.n

Core switching computer, namely core router or Layer 3 switch

4.1 - 4.n

terminal

5

access area

6.1 - 6. n

(Access) switch

7

subnet

8th

Trunk or backbone, e.g. Fiber Optic cable

Claims (20)

Method for transmitting data in a network (1) via packet-switched end-to-end connections, according to which user data representing a file or a data stream is provided in the form of data packets from a terminal device (4.1 - 4.n) of the plurality of terminals (4.1 - 4.n) comprehensive network (1) via a core network (2) with at least two core switching computers (3.1 - 3.n) to another, the payload data receiving terminal (4.1 - 4.n) are transmitted, which is the terminal equipment (4.1 - 4.n) is in each case a network-capable terminal or a subnetwork (7), and the core switching computers (3.1 - 3.n) are based on a terminal device (4.1 - 4.n) contained in a respective data packet ) determine a path through the network (1) of a respective end-to-end connection and forward the data packets to the receiving terminal (4.1 - 4.n) via this path, thereby marking et that the payload of a file or a data stream is routed through one or more of the at least two core switching computers (3.1 - 3.n), but each data packet passes only one of the core switching computers (3.1 - 3.n), all to the network (1 ) are connected redundantly to all the central switching computers (3.1 - 3.n) and all the central switching computers (3.1 - 3.n) constantly exchange operating data with each other, which comprise at least their respectively current dynamic routing tables. A method according to claim 1, characterized in that the total in the network (1) resulting from the transmission of user data traffic is dynamically divided into the core switching computer (3.1 - 3.n), wherein the division by units on or in the respective user data emitting terminal (4.1 - 4.n) is controlled. Method according to Claim 2, characterized in that the distribution of the data traffic is controlled in accordance with the principle of load balancing by means of a software routine which is processed by units on or in the user data terminal equipment (4.1 - 4.n). Method according to one of claims 1 to 3, characterized in that the data are forwarded depending on their nature by the core switching computers (3.1 - 3.n) with different priority. A method according to claim 4, characterized in that the prioritization is carried out by the use of quality of service mechanisms. Method according to one of Claims 1 to 5, in which subnetworks (7) with components of different system areas are connected as terminal equipment (4.1 - 4.n) to the central switching computers (3.1 - 3.n) of the network (1) within a technical system, characterized in that the payload data in the network (1) are transmitted logically separated from each other by the same physical paths but with respect to the different system areas by defining virtual networks. A method according to claim 6, characterized in that the user data is transmitted encrypted between selected virtual networks. Method according to one of Claims 1 to 6, characterized in that all user data transmitted in the network (1) are transmitted in encrypted form. Network (1) for the packet-switched transmission of data via end-to-end connections, which comprises a core network (2), an access area (5) surrounding the core network (2) and a plurality via the access area (5) to the core network (5). 2) connected terminal devices (4.1 - 4.n), wherein it is in the terminal devices (4.1 - 4.n) each a network-capable terminal or a subnet (7) and in the core network (2) at least two as high performance units trained core switching computers (3.1 - 3.n) are arranged, characterized in that all terminal devices (4.1 - 4.n) at least during the existence of an involving end-to-end connection to all of the at least two core switching computers (3.1 -. n) of the core network (2) are connected and all the central switching computers (3.1 - 3.n) are intermeshed with each other, between them one or more exclusively the exchange of operating data, in particular the current dynamic R outingtables serving high performance compounds (8). Network (1) according to Claim 9, characterized in that one or more of the core switching computers (3.1 - 3.n) are Layer 3 switches. Network (1) according to claim 9 or 10, characterized in that one or more of the core exchange computers (3.1 - 3.n) are core routers. Network (1) according to claim 10 or 11, characterized in that the core switching computers (3.1 - 3.n) are equipped with a firewall. Network (1) according to one of Claims 9 to 12, characterized in that all the central switching computers (3.1 - 3.n) are interconnected by one or more optical fiber cables (8) for exchanging the operating data. Network (1) according to one of claims 9 to 13, characterized in that one or more terminal devices (4.1 - 4.n) are arranged in the access area (5) switches (6.1 - 6.n) to at least two core exchange computers (3.1 - 3.n) are connected. Network (1) according to one of Claims 9 to 14, characterized in that at least two terminal devices (4.1 - 4.n) are connected to the central switching computers (3.1 - 3.n) with the interposition of means for data encryption and decryption. Network (1) according to one of claims 9 to 15, characterized in that in the network (1) subnets with at least two terminal devices (4.1 - 4.n) and at least two core switching computers (3.1 - 3.n) are formed with the remaining network parts are each connected via a security gateway. Network (1) according to one of claims 9 to 16, wherein the network (1) is implemented on board a ship and the terminal devices (4.1 - 4.n) are individual devices or navigation devices arranged in a subnet (7) of ship automation and communication, sensors and effectors, and devices for monitoring and controlling the aforesaid components, characterized in that the core switching computers (3.1 - 3.n) are arranged at different sections of the ship with respect to the ship's longitudinal extension and at least two each connected along a ship's side along connecting lines (8) are interconnected. Network (1) according to claim 17, characterized in that a core exchange computer (3.1) in the bow area and a core exchange computer (3.2) is arranged in the rear area of the ship. Network (1) according to claim 17 or 18, characterized in that the terminal devices (4.1 - 4.n) comprise components of the onboard air conditioning. A network (1) according to any one of claims 17 to 19, wherein the network is implemented on board a ship of the navy, characterized in that the terminal means (4.1 - 4.n) comprise components of an on-board armament.

Images